In a landmark decision that will significantly impact the artificial intelligence (AI) landscape, the UK Supreme Court has ruled unanimously that an ‘inventor’ under the Patents Act 1977 must be a ‘natural person’ and not AI.
This ruling came as a conclusion to the appeal case Thaler v Comptroller-General of Patents, Designs and Trade Marks [2023], which had previously been dismissed in both the High Court and the Court of Appeal in an earlier session in 2021.
The appellant, Dr. Stephen Thaler, had, in 2018, filed for patents under the 1977 Act, listing an AI machine called DABUS as the inventor. Dr. Thaler argued his right to be granted patents stemmed from his ownership of DABUS. The application was however rejected by the Comptroller-General of Patents, leading to further legal action.
Essentially, the appeal brought forward two central issues. Firstly, the matter of whether AI can be labeled an inventor under the 1977 Act, and secondly, whether being the owner of the mentioned AI-machine entitled Dr. Thaler to apply for patents.
After interpreting several case law authorities to define an ‘inventor’, the court concluded that it has to be a ‘natural person’ who contributes to the inventive idea. Such a definition could not therefore extend to machines or AI. In this context, the court highlighted that the Patents Act 1977 and subsequent legislation were designed without anticipating machines acting as inventors.
Additionally, the court rebuffed the appellant’s claim for a de facto right to patents arising from his ownership of DABUS. Dr. Thaler’s argument hinged on the ‘doctrine of accession’. Rejecting this, the court stated that the doctrine of accession could not be applicable, given that DABUS was not the inventor and an invention cannot be ‘transferred’ to the owner of the AI that produced it.
This pivotal decision ensues amidst new developments in AI regulations in the EU’s AI Act and the world’s first AI Safety Summit which was held in the UK in November. The EU reached a provisional agreement for the AI Act in early December 2023, which is the world’s first artificial intelligence regulation. The legislation is aimed towards developing a ‘trustworthy AI ecosystem’.